Zhou et al. BMC Plant Biology (2018) 18:342 https://doi.org/10.1186/s12870-018-1536-6

RESEARCHARTICLE Open Access Foreign cry1Ac gene integration and endogenous borer stress-related genes synergistically improve insect resistance in sugarcane Dinggang Zhou1,2†, Xiaolan Liu1,2†, Shiwu Gao1, Jinlong Guo1, Yachun Su1, Hui Ling1, Chunfeng Wang1, Zhu Li1, Liping Xu1* and Youxiong Que1*

Abstract Background: Sugarcane (Saccharum spp. hybrids) is considered the most globally important sugar-producing crop and raw material for biofuel. Insect attack is a major issue in sugarcane cultivation, resulting in yield losses and sucrose content reductions. Stem borer (Diatraea saccharalis F.) causes serious yield losses in sugarcane worldwide. However, insect-resistant germplasms for sugarcane are not available in any collections all over the world, and the molecular mechanism of insect resistance has not been elucidated. In this study, cry1Ac transgenic sugarcane lines were obtained and the biological characteristics and transgene dosage effect were investigated and a global exploration of gene expression by transcriptome analysis was performed. Results: The transgene copies of foreign cry1Ac were variable and random. The correlation between the cry1Ac protein and cry1Ac gene copies differed between the transgenic lines from FN15 and ROC22. The medium copy lines from FN15 showed a significant linear relationship, while ROC22 showed no definite dosage effect. The transgenic lines with medium copies of cry1Ac showed an elite phenotype. Transcriptome analysis by RNA sequencing indicated that up/down regulated differentially expressed genes were abundant among the cry1Ac sugarcane lines and the receptor variety. Foreign cry1Ac gene and endogenous borer stress-related genes may have a synergistic effect. Three lines, namely, A1, A5, and A6, were selected for their excellent stem borer resistance and phenotypic traits and are expected to be used directly as cultivars or crossing parents for sugarcane borer resistance breeding. Conclusions: Cry1Ac gene integration dramatically improved sugarcane insect resistance. The elite transgenic offspring contained medium transgene copies. Foreign cry1Ac gene integration and endogenous borer stress- related genes may have a synergistic effect on sugarcane insect resistance improvement. Keywords: Sugarcane, cry1Ac gene, Insect resistance, RNA-Seq, Transcriptome analysis, Phenotypic traits

* Correspondence: [email protected]; [email protected] †Dinggang Zhou and Xiaolan Liu contributed equally to this work. 1Key Laboratory of Sugarcane Biology and Genetic Breeding, Fujian Agriculture and Forestry University, Ministry of Agriculture, Fuzhou 350002, Fujian, China Full list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Zhou et al. BMC Plant Biology (2018) 18:342 Page 2 of 14

Background number of the foreign cry1Ac gene on its expression Sugarcane (Saccharum spp. hybrids) is considered the level in transgenic sugarcane via particle bombardment, most important crop for sugar production globally and is apart from the findings of Joyce et al. [23], who sug- a valued raw material for the biofuel industry [1, 2]. Insect gested that transgene copy number does not influence attack is a major issue in sugarcane cultivation, resulting the gene expression in transgene lines. in yield losses and sucrose content reduction [2]. One sig- The objective of breeding is to obtain excellent agro- nificant sugarcane pest is stem borer (Diatraea saccharalis nomic characters or to retain the agronomic characteris- F., Lepidoptera, Crambridae), which affects sugarcane tics of the parental cultivar [24]. Thus, assessing the throughout the entire growing season and causes serious agronomic performance of transgenic sugarcane under yield losses of nearly 25–30% [3, 4]. However, insect-resist- field conditions is necessary in order to select excellent ant sugarcane germplasms are not available in any collec- transgenic offspring [23]. Arencibia et al. [25]werethe tions [3]. In addition, modern sugarcane cultivars are first to perform field trials of five insect-resistant trans- highly complex polyploid-aneuploids, with chromosome genic sugarcane events and they found that most of the numbers ranging from 80 to 130 [5]. It is thus almost im- transgenic lines had agronomic traits similar to that of the possible to breed an insect-resistant variety by means of receptor variety. Gao et al. [18] tested the field perform- traditional cross-breeding. Genetic engineering is ex- ance of transgenic cry1Ac sugarcane and discovered that pected to play an important role in improving the insect these lines exhibited better phenotypic traits than the resistance of sugarcane [6–8] and could facilitate the de- non-transgenic sugarcane. Wang et al. [19] investigated velopment of insect-resistant sugarcane varieties or germ- the field performance of five single-copy transgenic sugar- plasms for use in cross-breeding. cane lines, which exhibited excellent cane borer resistance To date, the cry gene has been effectively used to con- and herbicide tolerance but poor agronomic traits. trol stem-borer pests in many crops, including rice The genetic engineering of plants with enhanced toler- (Oryza sativa)[9], corn (Zea mays)[10], cotton (Gossy- ance to biotic stresses (i.e., insect stress) typically in- pium hirsutum)[11], potato (Solanum tuberosum)[12] volves complex multigene networks and may therefore and soybean (Glycine max)[13], which have widespread have the potential to introduce unintended effects due commercial applications and verified safety [14–16]. to the location effect of transgene integration [26]. How- There is an urgent need for borer resistance traits, and a ever, the cry1Ac transgenic sugarcane insertion site and series of studies that introduced the cry gene into sugar- its flanking sequence are complex, and the effects of par- cane successfully obtained insect-resistant sugarcane ticle bombardment remain unclear. The growth and de- lines [1, 4, 17–19]. However, only one case involving velopment of plants, including transgenic plants, along insect-resistant transgenic sugarcane has been approved with plant responses to abiotic and biotic stresses, in- for commercial planting in Brazil. The first report on volves a complex network of gene regulation [27]. In re- insect-resistant sugarcane lines based on the use of the cent years, there has been an increasing number of cry1Ab gene [20]. Recently, Gao et al. [18] successfully reports on gene expression analysis in plants during de- introduced the cry1Ac gene into sugarcane and obtained velopment and stress using a global transcriptomic ap- insect- resistant transgenic lines. Wang et al. [19] suc- proach [27]. However, these reports mainly focused on cessfully introduced the cry1Ab and EPSPS genes to- model plants due to their available genome sequence in- gether and obtained transgenic sugarcane lines with formation. Many studies have investigated the transcrip- insect resistance and herbicide tolerance. tomes of non-model plants, including sugarcane, for The influence of transgene copy number on gene ex- which the sequencing of the whole genome is currently pression levels is complex [21]. The gene balance hy- in progress [27]. Microarrays, serial analysis of gene pothesis suggests that increasing the transgene copy expression and RNA sequencing (RNA-Seq) are the number would upregulate gene expression levels, and three most commonly used methods for transcriptome thus a correlation (positive or negative) must exist be- analysis in gene expression studies [28, 29]. RNA-Seq, tween gene copy number and gene expression level [22]. which allows for the near-complete characterization of Therefore, transcript abundance must increase with gene transcriptomic events occurring in a specific tissue at a dosage in order to increase protein abundance [22]. certain time, has been widely applied and proven par- However, low copy-number exogenous genes are consid- ticularly useful in non-model plants, including sugarcane ered to be beneficial for plant improvement, particularly [28, 30–32]. For transgenic plants, a number of studies in diploid plants [21]. It is generally considered that a to date have used transcriptome analysis to study gene low gene copy number would decrease the possibility of expression or assess the impact of genetic engineering transgene co-suppression, while multiple gene copies [29, 33–36]. Misra et al. [35] identified differentially may result in gene silencing and co-suppression [21]. To expressed genes (DEGs) in AtMYB12-expressing trans- date, very little is known about the influence of the copy genic tobacco lines by microarray transcriptome analysis. Zhou et al. BMC Plant Biology (2018) 18:342 Page 3 of 14

Nietzsche et al. [36] employed a global transcriptional The CYC gene was selected for transgene copies esti- profiling approach using microarray to identify tran- mation and the transgene cry1Ac copies in transgenic scriptional changes in 35S-STKR1 Arabidopsis in order sugarcane are shown in Table 1. The proportions of dif- to interpret the observed phenotypic and metabolic ferent transgene cry1Ac copies in transgenic sugarcane changes. Cai et al. [37] employed RNA-seq to identify are shown in Fig. 1. The cycle threshold (Ct) value of transcriptional changes in transgenic ZmWRKY17 the endogenous reference gene CYC ranged from Arabidopsis to reveal salt stress and abscisic acid (ABA) 23.837 ± 0.082~ 25.398 ± 0.029, and the mean Ct was responsive genes in the ABA signaling pathway. Chung 24.590 ± 0.076. Variance analysis indicated that there et al. [38] performed RNA-Seq to identify the direct tar- was no significant difference between the Ct value of all get genes of the OsNAC proteins in transgenic OsNAC of the transgenic and control lines. rice and to elucidate the molecular regulatory networks For Group I (the transgenic lines from the receptor var- of the root architectures of RCc3:OsNACs for drought iety FN15), the copy number of the cry1Ac gene per single tolerance. Based on all the above, a global transcriptome cell ranged from 0.03 ± 0.01 copies~ 28.64 ± 0.40 copies, analysis in cry1Ac transgenic sugarcane could help iden- of which the Ct value of the corresponding line was tify DEGs and elucidate the selected or the foreign gene 35.987 ± 0.323 (B1) and 26.036 ± 0.02 (A5), respectively. networks that are associated with the performance of For Group II (the transgenic lines from the receptor var- sugarcane traits, particularly insect resistance. iety ROC22), the copy number of the cry1Ac gene per sin- To improve the insect resistance of sugarcane, the gle cell ranged from 1.59 ± 0.04 copies~ 63.47 ± 0.33 cry1Ac gene was genetically engineered via particle bom- copies, of which the Ct value of the corresponding line bardment in our previous work [2, 17, 18, 39]. In the was 29.87 ± 0.04 (D1) and 24.37 ± 0.01 (D4), respectively. present study, in order to elucidate how foreign cry1Ac Further analysis of the cry1Ac gene copies showed gene integration and endogenous borer-stress-related that 0–1 copies/2C accounted for 18%, while 1–16 genes contribute to insect resistance improvement in (=16) copies/2C, 17–32 copies/2C, and > 33 copies/2C sugarcane, and to select elite insect-resistant cry1Ac accounted for 23, 41, and 18%, respectively. For Group transgenic sugarcane lines, the performance and molecu- I, 0–1 copies/2C accounted for 27%, while 1–16 (=16) lar characteristics of these lines, including the stalk borer copies/2C and 17–32 copies/2C accounted for 9 and damage level, the main agronomic traits and the correl- 64%, respectively. No line possessed a copy number of ation between cry1Ac gene copy number and gene ex- > 31 copies/2C. For Group II, 1–16 (=16) copies/2C pression level, were investigated. The expression of the and > 31 copies/2C accounted for 50 and 50%, respect- foreign cry1Ac gene and DEGs were then analyzed using ively. No line possessed a copy number of 0–1copies/ a global transcriptome approach by high-throughput 2C and 11–30 copies/2C. RNA-Seq. Furthermore, the synergistic effects of the for- The results suggest that the copies of the exogenous eign cry1Ac gene and endogenous borer-stress related cry1Ac gene in the transgenic sugarcane from particle genes, and those genes identified from the DEGs associ- bombardment are variable and random.. ated with borer resistance metabolism in sugarcane, were also discussed. The present study provides novel Transgene expression showed varied relationships with insights into the mechanisms of foreign cry1Ac gene in- transgene copies tegration and endogenous borer stress-related genes in The expression level of the cry1Ac protein in trans- insect resistance improvement in sugarcane. genic sugarcane is summarized in Table 2.Whetherthe transgenic lines were from the receptor FN15 or ROC22, significant differences were found among all Results transgenic lines (except B1 and B2) and between the Transgene copies via particle bombardment are variable transgenic and the control line. Among the Group I and random transgenic lines, the line with the highest cry1Ac pro- − To estimate the copy number of the foreign cry1Ac gene, tein expression was A3 with 547.45 ± 0.06 ng·g 1,while a quantitative TaqMan real-time PCR method was estab- fortheGroupII,K5hadthehighestcry1Acproteinex- − lished. This method was based on double-standard pression level of 113.42 ± 0.24 ng·g 1. curves of cry1Ac and CYC/APRT/P4H gene, which were To explore the correlation between the cry1Ac pro- integrated into the multi-target recombined plasmid tein and cry1Ac gene copies of the transgenic sugar- pG1AcAPC0229 (p1AAPC). We discovered that, for cane, a scatter plot was drawn (Fig. 2). For Group I, transgene copies estimation in sugarcane, the internal especially, the A1–A6 lines with medium copies reference genes CYC, APRT, and P4H did not differ sig- showed a significant linear relationship (P-value and R nificantly, obtaining a high amplification efficiency be- of the Pearson’s correlation analysis were 0.002 and tween 0.95 and 1.05 [2]. 0.818, respectively), and the higher the cry1Ac gene Zhou et al. BMC Plant Biology (2018) 18:342 Page 4 of 14

Table 1 Transgene copies of cry1Ac transgenic sugarcane by quantitative TaqMan real-time PCR Lines cry1Ac Ct value cry1Ac copies CYC Ct value CYC copies Relative copies (cry1Ac/CYC) Mean ± SE Group I A1 26.25 ± 0.05 24.83 ± 0.77 24.10 ± 0.05a 54.23 ± 1.7 0.46 ± 0.02 A2 26.21 ± 0.02 25.46 ± 0.20 24.10 ± 0.08a 51.79 ± 0.55 0.49 ± 0.01 A3 26.13 ± 0.03 26.88 ± 0.64 24.37 ± 0.10a 45.12 ± 3.13 0.60 ± 0.02 A4 26.24 ± 0.07 24.87 ± 1.20 24.37 ± 0.01a 45.16 ± 0.43 0.55 ± 0.05 A5 26.04 ± 0.02 28.64 ± 0.40 24.28 ± 0.05a 48.02 ± 1.66 0.6 ± 0.01 A6 26.19 ± 0.07 25.75 ± 1.27 24.52 ± 0.05a 40.62 ± 1.25 0.63 ± 0.05 B1 35.99 ± 0.32 0.03 ± 0.01 24.45 ± 0.05a 42.66 ± 1.35 0.00 ± 0.00 B2 34.94 ± 0.11 0.04 ± 0.00 24.33 ± 0.04a 46.35 ± 1.13 0.00 ± 0.00 B4 35.07 ± 0.11 0.04 ± 0.00 23.84 ± 0.08a 65.22 ± 3.59 0.00 ± 0.00 I2 27.59 ± 0.06 8.05 ± 0.32 24.70 ± 0.06a 35.92 ± 1.47 0.22 ± 0.02 I4 26.00 ± 0.03 24.63 ± 0.57 24.76 ± 0.08a 34.48 ± 1.96 0.71 ± 0.03 Group II D1 29.87 ± 0.04 1.59 ± 0.04 25.50 ± 0.01a 20.7 ± 0.07 0.08 ± 0.00 D2 25.24 ± 0.06 35.60 ± 1.50 25.40 ± 0.03a 22.14 ± 0.44 1.61 ± 0.12 D4 24.37 ± 0.01 63.47 ± 0.33 24.86 ± 0.01a 32.01 ± 0.19 1.98 ± 0.02 K2 25.07 ± 0.01 39.78 ± 0.13 24.81 ± 0.02a 33.26 ± 0.35 2.39 ± 0.01 K3 28.75 ± 0.027 3.37 ± 0.06 24.50 ± 0.03a 41.2 ± 0.91 0.08 ± 0.00 K5 28.48 ± 0.02 4.04 ± 0.06 24.89 ± 0.07a 31.45 ± 1.43 0.13 ± 0.00 Control FN15 ––24.60 ± 0.05a 38.54 ± 1.24 – ROC22 ––24.51 ± 0.03a 40.96 ± 0.87 –

ddH2O ––––– Note: Mean Mean value of three replicates, SE Standard error, “-” means undetected

copies, the higher the cry1Ac protein expression. Cry1Ac transgenic sugarcane exhibited excellent cane Nevertheless, not all the medium-copy lines conformed borer resistance to this linear relationship. For instance, the medium To investigate the insect resistance efficiency, the stalk copy line I4 had a cry1Ac protein content of only 6.1 ± borer damage level of the 17 cry1Ac transgenic lines and − 0.05 ng·g 1 leaf. In Group II, both the “high-copy” line two receptor varieties FN15 and ROC22 are shown in D4 and “medium-copy” line K2 showed lower cry1Ac Table 3. protein expression levels. For Group II, the P-value and The results indicated that the borer damage percent- R were 0.589 and − 0.282, respectively, which showed age of the receptor varieties FN15 and ROC22 was as that no definite linear relationship existed between the high as 85.00 ± 2.89 and 93.33 ± 3.33. For Group I, the cry1Ac protein and cry1Ac copies. borer damage ratio of all cry1Ac lines was significantly

Fig. 1 Proportion of different copy numbers of the cry1Ac gene in transgenic sugarcane. a Proportion of cry1Ac gene copy numbers in both FN15 and ROC22 transgenic sugarcane; b Proportion of cry1Ac gene copy numbers in FN15 transgenic sugarcane; c Proportion of cry1Ac gene copy numbers in ROC22 transgenic sugarcane Zhou et al. BMC Plant Biology (2018) 18:342 Page 5 of 14

Table 2 Cry1Ac protein content in sugarcane leaves detected the insect resistance effect of different receptor geno- by quantitative ELISA types was variable. Lines Cry1Ac protein Lines Cry1Ac protein −1 −1 (ng·g , Mean ± SE) (ng·g , Mean ± SE) Phenotypic trait performance of cry1Ac transgenic f i Group I A1 445.79 ± 0.15 B1 19.32 ± 0.18 sugarcane demonstrated commercial potential A2 451.90 ± 0.24e B2 19.45 ± 0.01i Seventeen transgenic lines and two corresponding recep- A3 547.45 ± 0.06a B4 42.18 ± 0.21h tor varieties FN15 and ROC22, were subjected to major A4 468.47 ± 0.02d I2 90.99 ± 0.16g agronomic traits analysis (shown in Table 4). In Group I, the plant height of the transgenic line A1 A5 501.78 ± 0.25b I4 6.10 ± 0.05j c (233.11 ± 8.76 cm) was significantly higher than that of A6 469.19 ± 0.08 FN15 (209.92 ± 6.35 cm), while those of lines A2, A3, Control variety FN15 0.00 ± 0.05 k A4, A5 and A6 did not differ significantly from that of Group II D1 102.56 ± 0.05b K2 36.17 ± 0.23f FN15, ranging from196.89 ± 4.87 cm to 214.50 ± 3.89 cm. D2 87.04 ± 0.35c K3 24.23 ± 0.12e The remaining lines of Group I were significantly lower D4 67.3 ± 0.28d K5 113.42 ± 0.24a than that of FN15, with the height of line I2 being the smallest with an average height of only 85.00 ± 7.83 cm. Control variety ROC22 0.00 ± 0.03g In Group II, all of the lines had significantly lower Note: Lowercase in the column followed by the same letters mean no significant difference at P = 0.05 level to their corresponding receptor variety heights than that of ROC22, except for lines K2 and K3. Stem diameter is an important agronomic trait in sugar- lower than that of the receptor variety FN15. In Group cane. In Group I, the stem diameter of A1, A5, and A6 I, the damage ratio in lines A1–A6 was significantly was similar to that of FN15 (no significant difference); lower than in B1, B2, B4, I2, and I4, and the line with however, the remaining transgenic lines in this group the lowest borer damage ratio was A5 at only 8.33 ± were significantly lower than that of FN15. In Group II, 1.67. For the Group II transgenic lines, the borer damage all of the lines had significantly lower stem diameters ratio in all of the lines was lower than that of the recep- than that of ROC22. The Brix value reflects the sucrose tor variety ROC22. These results indicated that the in- content of sugarcane. The Brix value of most of the sect resistance efficiency of the cry1Ac transgenic lines transgenic lines in Group I was equivalent to that of of the receptor variety FN15 was better than that of FN15, while the A5, B1, B4, and I2 lines were signifi- ROC22. In addition, variance analysis indicated that cantly higher than that of FN15. The probable reason is there was no significant difference in the borer damage that stem borer damage significantly reduced the Brix ratio between the receptor varieties FN15 and ROC22. value and sucrose content. The increased borer resist- These findings showed that the introduction of the ance of line A5 led to reduced damage and thus a higher exogenous cry1Ac gene significantly improved the stem Brix value, whereas the higher Brix values in lines B1 borer resistance of the transgenic sugarcane lines and I2 could be due to the decreased height and reduced (Additional file 1: Figure S1, A1 line as an example), but diameter, which renders the storage capacity smaller and

Fig. 2 Correlation between cry1Ac gene copies and cry1Ac protein content in transgenic sugarcane. a Scatter diagram of the cry1Ac protein vs. cry1Ac gene copies for 11 cry1Ac transgenic lines from the receptor variety FN15: A1, A2, A3, A4, A5, A6, B1, B2, B4, I2, and I4; the P-value and R value from the Pearson’s correlation analysis were 0.002, 0.818, respectively. b Scatter diagram of the cry1Ac protein vs. cry1Ac gene copies for six cry1Ac transgenic lines from the receptor variety ROC22: D1, D2, D4, K2, K3 and K5. The P-value and R value from the Pearson’s correlation analysis were 0.589 and − 0.282, respectively Zhou et al. BMC Plant Biology (2018) 18:342 Page 6 of 14

Table 3 Variance analysis of borer damage ratio of cry1Ac transgenic sugarcane lines Lines The borer damage percentage Lines The borer damage percentage (%, Mean ± SE) (%, Mean ± SE) Group I A1 15.00 ± 2.89e B1 36.67 ± 3.33 c,d A2 16.67 ± 1.67e B2 40.00 ± 0.00 c,d A3 11.67 ± 6.01e B4 33.33 ± 6.67d A4 10.00 ± 0.00e I2 46.67 ± 3.33b,c A5 8.33 ± 1.67e I4 53.33 ± 3.33b A6 20.00 ± 5.00e Control variety FN15 85.00 ± 2.89a Group II D1 31.67 ± 1.67c K2 36.67 ± 4.41b,c D2 33.33 ± 1.67c K3 43.33 ± 1.67b D4 30.00 ± 2.89c K5 33.33 ± 1.67c Control variety ROC22 93.33 ± 3.33a Note: Lowercase in the column followed by the same letters mean no significant difference at P = 0.05 level to their corresponding receptor variety thereby increases the sucrose concentration. In Group in the transgenic lines of Group II, except for line K2, II, the Brix value of each line was significantly lower which was significantly lower than that of ROC22. than that of ROC22. As an important index of sugarcane yield, the number of productive tillers of lines B2, B4, I2, The theoretical sucrose yields varied in the cry1Ac and I4 per block (31.2 m2) was significantly lower than transgenic sugarcane lines that of FN15, while the other lines in this group were Sugarcane yield and theoretical sucrose production are equivalent to that of FN15. The same pattern was found the two most important indexes of sugarcane as a crop.

Table 4 Performance of the major agronomic traits of cry1Ac transgenic sugarcane lines Lines Diameter/cm Height /cm Brix values/% Number of Sucrose Weight per stem Sugarcane Theoretical millable content/% Kg/stem yield t/ha sugar yield t/ha stalks/block Group I A1 3.14 ± 0.09a,b 233.1 ± 8.8a 21.48 ± 0.57c,d 240.0 ± 19.1a 15.22 ± 0.61c,d 1.80 ± 0.10a 138.79 ± 7.80a 21.13 ± 1.19a A2 2.99 ± 0.02b,c 214.5 ± 3.9b 20.71 ± 0.15d 216.0 ± 5.8a,b,c 14.72 ± 0.16d 1.51 ± 0.03c,d 104.22 ± 2.04c 15.34 ± 0.30d A3 2.80 ± 0.04 c,d 205.8 ± 2.6b 20.61 ± 0.09d 222.7 ± 8.9a,b 14.61 ± 0.10d 1.27 ± 0.03e 90.39 ± 2.17d 13.20 ± 0.32e A4 2.97 ± 0.07b,c 196.9 ± 4.9b 22.09 ± 0.11a,b,c 207.6 ± 11.4a,b,c 16.21 ± 0.12a,b,c 1.36 ± 0.04d,e 90.71 ± 2.61d 14.70 ± 0.42d A5 3.12 ± 0.01a,b 213.7 ± 6.8b 22.55 ± 0.16a 220.0 ± 11.1a,b 16.71 ± 0.17a 1.63 ± 0.04b,c 115.13 ± 3.07b 19.24 ± 0.51b A6 3.08 ± 0.04a,b 203.9 ± 7.4b 22.04 ± 0.19a,b,c 217.2 ± 6.2a,b,c 16.16 ± 0.21ab,c 1.52 ± 0.05c 105.68 ± 3.53b,c 17.07 ± 0.57c B1 2.42 ± 0.02f 116.3 ± 2.1d 23.07 ± 0.15a 181.6 ± 18.4c,d 17.27 ± 0.16a 0.53 ± 0.00g,h 31.13 ± 0.16e,f 5.38 ± 0.03 f,g B2 2.42 ± 0.09f 126.5 ± 5.5d 22.4 ± 0.21a,b 100.8 ± 6.4f,g,h 16.55 ± 0.23a,b 0.58 ± 0.05f,g,h 18.79 ± 1.62g 3.11 ± 0.27i B4 2.38 ± 0.06f 154.5 ± 9.3c 22.53 ± 0.19a 144.0 ± 6.0e 16.69 ± 0.20a 0.69 ± 0.07f,g 31.71 ± 3.24e,f 5.29 ± 0.54f,g I2 2.56 ± 0.08e,f 85.0 ± 7.83e 22.60 ± 0.35a 134.4 ± 13.8e,f 16.76 ± 0.38a 0.44 ± 0.03h 18.84 ± 1.33g 3.16 ± 0.22i I4 2.82 ± 0.10 c,d 113.3 ± 4.7d 22.3 ± 0.42a,b 134.4 ± 12.4e,f,g 16.44 ± 0.45a,b 0.71 ± 0.05f 30.4 ± 2.21e,f 5.00 ± 0.36g,h Control FN15 3.25 ± 0.05a 209.9 ± 6.4b 21.47 ± 0.67b,c,d 206.4 ± 7.6a,b,c 15.54 ± 0.72b,c,d 1.74 ± 0.08a,b 115.15 ± 5.09b 17.89 ± 0.79b,c variety Group II D1 2.22 ± 0.03b,c 133.1 ± 7.3e 22.08 ± 0.17b,c 203.0 ± 6.7a,b,c 16.20 ± 0.18b,c 0.52 ± 0.01c 33.6 ± 0.92d 5.44 ± 0.15e,f D2 2.43 ± 0.08b 183.8 ± 8.6c,d 20.02 ± 0.31e 199.5 ± 3.6a,b,c 13.97 ± 0.33e 0.85 ± 0.08b 54.31 ± 5.34b 7.58 ± 0.75b,c,d D4 2.34 ± 0.08b,c 176.7 ± 5.5 c,d 21.77 ± 0.17b,c,d 228.0 ± 12.9a 15.86 ± 0.18b,c,d 0.76 ± 0.04b,c 55.45 ± 3.04b 8.79 ± 0.48b K2 2.24 ± 0.07b,c 222.2 ± 6.5a 21.57 ± 0.38b,c,d 182.0 ± 13.5c 15.64 ± 0.42b,c,d 0.88 ± 0.07b 51.20 ± 4.31b,c 8.01 ± 0.67b,c K3 2.24 ± 0.05b,c 214.4 ± 7.6a 20.73 ± 0.43d,e 150.0 ± 9.3d 14.74 ± 0.47d,e 0.84 ± 0.04b 40.42 ± 1.94 c,d 5.96 ± 0.29d,e,f K5 2.29 ± 0.06b,c 191.2 ± 7.9b,c 20.97 ± 0.28c,d,e 195.0 ± 5.6b,c 14.99 ± 0.31c,d,e 0.79 ± 0.02b 49.11 ± 1.17b,c 7.36 ± 0.18b,c,d,e Control ROC22 2.71 ± 0.07a 216.8 ± 6.5a 23.40 ± 0.20a 216.0 ± 5.3a,b 17.63 ± 0.22a 1.25 ± 0.09a 86.54 ± 6.29a 15.26 ± 1.11a variety Note: Lowercase in the column followed by the same letters mean no significant difference at P = 0.05 level to their corresponding receptor variety Zhou et al. BMC Plant Biology (2018) 18:342 Page 7 of 14

The sugarcane yield and theoretical sucrose yield results (phenylalanine ammonia lyase) are involved in anti-insect estimated from the plot survey are shown in Table 4. related signaling pathways [40, 41]. For transgenic line A1, the sugarcane yield and theoret- In addition to these anti-insect metabolic pathways, − ical sucrose production (138.79 ± 7.80 t·ha 1 and 21.13 ± some basic metabolic pathways, such as starch and su- − 1.19 t·ha 1) were significantly higher than that of FN15 crose metabolism, galactose metabolism (carbohydrate − − (115.15 ± 5.09 t·ha 1 and 17.89 ± 0.79 t·ha 1). Further- metabolism), and photosynthesis-antenna proteins (en- more, the sugarcane yield and theoretical sucrose pro- ergy metabolism), were also altered. These are antici- duction of lines A5 and A6 were equivalent to that of pated to affect the growth and development of plants, FN15, while those of the remaining lines were signifi- which would probably result in the poor performance of cantly lower than that of FN15 in this group. However, agronomic traits (i.e., short and slender stalks) of the all of the transgenic lines of Group II were significantly transgenic line B4, while no effect on the growth and de- lower than that of ROC22. velopment of lines A1 and A5 was observed.

Transcriptome dynamics in the cry1Ac transgenic The expression of endogenous borer-stress responsive genes sugarcane is demonstrated by RNA-Seq in cry1Ac transgenic sugarcane is differentially regulated Transcriptome analysis of the transgenic lines (A1, A5, To elucidate the molecular mechanisms of insect resist- and B4) and receptor variety FN15 was conducted using ance, the differential expression of endogenous borer-stress RNA-Seq technology. responsive genes was investigated. Illumina sequencing of the 12 samples obtained For Ca2+ channel-related proteins, group comparison 83.50 Gb clean reads with more than 95.48% Q20 analysis of the transgenic sugarcane and the control line bases and more than 90.67% Q30 bases for each (CK vs. A1) indicated that four DEGs encoding sample (see Additional file 2: Table S1). The Trinity calcium-dependent protein kinases (CDPKs, all four package assembled 65,995 unigenes, which was used as DEGs were up–regulated) were observed, while three the sugarcane reference library, with an average length of DEGs (all three DEGs up–regulated) in the group com- 715 bp and N50 length of 1049 bp, including 13,813 uni- parison of CK vs. A5, and zero DEGs in the group com- genes measuring more than 1 kb (Additional file 2:Figure parison CK vs. B4 were observed. Additionally, eight S2 and Table S2). A total of 65,995 unigenes were DEGs that encoding MAPKs (three up-regulated DEGs functionally annotated using the Non-redundant (Nr), and five down-regulated DEGs) were observed in group Swiss-Prot, EuKaryotic Orthologous Groups (KOG), and comparison CK vs. A1, while 14 DEGs (nine Kyoto Encyclopedia of Genes and Genomes (KEGG) up-regulated DEGs and five down-regulation DEGs) databases and the annotation statistics are shown in were observed in the group comparison of CK vs. A5, Additional file 2: Table S3. and 17 DEGs were observed in the group comparison of The DEG statistics are shown in Additional file 2: CK vs. B4 (all 17 DEGs were up–regulated) (Additional Figure S3. A total of 6675 DEGs (4744/1931 up−/down-- file 3: Table S4 and Fig. 4). regulated), 9181 DEGs (6410/2771 up−/down-regulated), For JA signaling pathway related genes, all the DEGs en- and 5050 DEGs (1330/3720 up−/down-regulated) were coding JAR1 (jasmonic acid resistant 1), JAZs and MYC2 identified in transgenic lines A1, A5, and B4 compared (a basic-helix-loop-helix transcription factor) were ob- with CK, respectively. The pathway enrichment analysis served to be up-regulated or down-regulated. All of the provided important information for investigating specific DEGs encoding these proteins in transgenic lines A1 and biological processes that could be influenced by the ex- A5 were up-regulated compared to the non-transgenic pression of the foreign cry1Ac gene in sugarcane. sugarcane, while all of the DEGs encoding JAZs and The large number of DEGs allowed for functional anno- MYC2 in the transgenic line B4 were down-regulated. For tation by KEGG enrichment analysis. Pathway enrichment the group comparison of CK vs. A1, two DEGs encoding analysis showed that the DEGs were mainly enriched in JAR1, 19 DEGs encoding JAZs, and one DEG encoding pathways associated with signal transduction, biosynthesis MYC2 were identified. Similarly, for the group compari- of secondary metabolism, environmental adaptation and son of CK vs. A5, two DEGs encoding JAR1, 19 DEGs en- lipid metabolism (Fig. 3). As shown in Fig. 3,planthor- coding JAZs, and one DEG encoding MYC2 were mone signal transduction, plant-pathogen interaction, identified. For the group comparison of CK vs. B4, though benzoxazinoid biosynthesis, phenylpropanoid biosyn- eight DEGs encoding JAZs and one DEG encoding MYC2 thesis, and flavonoid biosynthesis were also regulated. were identified, no DEG encoding JAR1 was observed. These results are similar to those of previous studies For benzoxazinoid biosynthesis-related genes, group whereby MAPKs (mitogen-activated protein kinases), Ca2 comparisons of CK vs. A1 indicated that four DEGs + channel proteins, ROS (reactive oxygen species), JAZs encoding hydroxamic acids (DIMBOA/DIBOA) biosyn- (jasmonate ZIM-domain proteins) and PAL thetic enzymes (two up-regulated DEGs and two Zhou et al. BMC Plant Biology (2018) 18:342 Page 8 of 14

NA 0.002 0.000 0.000 0.002 0.004 Phenylalanine metabolism class NA 0.002 0.010 0.009 0.041 0.823 Phenylalanine, tyrosine and tryptophan biosynthesis Amino acid metabolism NA 0.091 0.047 0.003 0.001 0.445 Benzoxazinoid biosynthesis 0.8 Biosynthesis of other secondary metabolites NA 0.199 0.342 NA 0.346 0.042 Flavone and flavonol biosynthesis Carbohydrate metabolism NA 0.004 0.077 0.011 0.047 0.000 Flavonoid biosynthesis 0.6 Energy metabolism 0.350 0.029 0.000 0.000 0.000 0.008 Phenylpropanoid biosynthesis NA 0.155 0.205 0.005 0.021 0.010 Stilbenoid, diarylheptanoid and gingerol biosynthesis Environmental adaptation NA 0.033 0.035 0.002 0.014 0.040 Amino sugar and nucleotide sugar metabolism 0.4 Global and overview maps 0.024 NA 0.617 NA 0.420 NA C5−Branched dibasic acid metabolism Glycan biosynthesis and metabolism NA 0.092 0.155 0.045 0.271 0.001 Galactose metabolism 0.2 Immune system NA 0.095 0.252 0.212 0.059 0.007 Starch and sucrose metabolism Lipid metabolism 0.000 0.616 1.000 0.109 0.998 0.832 Oxidative phosphorylation 0.05 0.004 0.097 0.019 0.611 0.826 0.165 Photosynthesis Membrane transport NA 0.959 0.368 0.845 0.692 0.000 Photosynthesis − antenna proteins Metabolism of cofactors and vitamins NA 0.857 0.684 0.974 0.850 0.030 Circadian rhythm − plant Metabolism of other amino acids NA 0.000 0.000 0.000 0.000 0.000 Plant−pathogen interaction Metabolism of terpenoids and polyketides 0.853 0.001 0.001 0.000 0.001 0.000 Biosynthesis of secondary metabolites Replication and repair 0.028 0.007 0.084 0.000 0.010 0.000 Metabolic pathways NA 0.027 0.066 NA NA NA Other types of O−glycan biosynthesis Signal transduction NA 0.249 0.031 0.821 0.918 0.494 Ether lipid metabolism 0.332 0.029 0.020 0.016 0.053 0.123 Glycerolipid metabolism NA 0.000 0.000 0.000 0.001 0.095 Glycerophospholipid metabolism NA 0.000 0.000 0.000 0.000 0.214 Linoleic acid metabolism NA 0.863 0.949 0.135 0.747 0.026 Steroid biosynthesis 0.219 0.000 0.000 0.000 0.000 0.008 alpha−Linolenic acid metabolism NA 0.007 0.020 NA 0.148 0.376 ABC transporters NA 0.734 0.062 0.979 0.864 0.000 Porphyrin and chlorophyll metabolism NA 0.394 0.256 NA 0.663 0.028 Riboflavin metabolism NA 0.071 0.008 0.946 0.713 0.093 Ubiquinone and other terpenoid−quinone biosynthesis NA 0.122 0.061 0.156 0.028 0.183 Cyanoamino acid metabolism NA 0.972 0.797 0.187 0.043 0.892 Glutathione metabolism NA 0.265 0.136 0.462 0.037 NA Brassinosteroid biosynthesis NA 0.017 0.098 0.042 0.058 0.006 Carotenoid biosynthesis NA 0.000 0.009 0.140 0.001 0.445 Diterpenoid biosynthesis NA 0.027 0.066 0.163 0.063 0.022 Zeatin biosynthesis 0.034 0.186 0.111 0.960 0.049 0.056 Mismatch repair 0.084 0.008 0.003 0.177 0.297 0.361 Non−homologous end−joining NA 0.163 0.155 0.007 0.041 0.585 Phosphatidylinositol signaling system 0.647 0.000 0.000 0.000 0.000 0.004 Plant hormone signal transduction class A1−VS−A5 A1−VS−B4 A5−VS−B4 CK−VS−A1 CK−VS−A5 CK−VS−B4

Fig. 3 Heat map of the enriched KEGG pathways for differentially expressed genes. Pathway enrichment analysis was used for all the identified differentially expressed genes detected by RNA-Seq. The color bar on the left represents the different KEGG classification annotations. Each row represents a pathway (at least one sample of P ≤ 0.05), and each column represents a comparison group down-regulated DEGs) were observed, while five DEGs Sugarcane borer damage causes great agricultural losses, (two up-regulated DEGs and two down-regulated) in the resulting in approximately 25–30% losses in cane yield group comparison of CK vs. A5, and only one and 15% reductions in sucrose content annually [3, 4, 50]. down-regulated DEG in the group comparison CK vs. B4 However, traditional cross-breeding for sugarcane insect were observed. (Additional file 3: Table S4 and Fig. 4). resistance is limited owing to both the lack of The findings of these DEG analyses suggest that en- insect-resistant germplasms and the complexity of the in- dogenous borer-stress related genes play a positive role heritance background [3, 5]. Prior work has documented in improving insect resistance in transgenic sugarcane the effectiveness of transgenic crops expressing Bt insecti- lines A1 and A5, but negatively influence line B4. cidal proteins. The cry1Ac protein has become an import- ant tool in pest management in crops [51]. As an industrial and a vegetative propagation crop, sugarcane is Discussion suitable for insect-resistant transgenic modification via the Controlling diseases and pests is necessary for achiev- importation of Bt genes such as cry1Ac. Therefore, several ing food security. The use of resistant varieties is transgenic sugarcane lines expressing the cry1Ac crystal widely considered as one of the most cost-effective protein were developed in our previous study [18]. The measures for achieving food security. Both traditional main purpose of the present study was to explore how for- cross breeding and modern genetic engineering breed- eign cry1Ac gene integration and endogenous borer-stress ing approaches have been employed to improve crop related genes improve insect resistance in sugarcane, as resistance, productivity, and other desirable traits. well as to screen for elite insect-resistant transgenic Transgenic breeding, in particular has tremendous po- cry1Ac lines. tential for the introduction of novel traits that are Simple transgene integration enables molecular not normally present in the plant genome, i.e., herbi- characterization [52]. Integration of a single and intact cide tolerance or insect resistance [42–49]. copy (or low copies, 1–2) of the transgene expression Zhou et al. BMC Plant Biology (2018) 18:342 Page 9 of 14

Fig. 4 A model process of transgenic sugarcane responses to insect attack and heat maps of endogenous borer-stress related differentially expressed genes. a A model process of transgenic sugarcane responses to insect attack; b Heat maps of endogenous borer-stress related differentially expressed genes. CDPKs, OPDAs, MAPKs, JAR1, JAZs and MYC2: induced defense proteins related to insect attack; DIMBOA/DIBOA: hydroxamic acids, direct defense proteins related to insect attack

cassette can reduce the potential for unintended inser- Ismail’s[1] findings whereby independent transgenic tional inactivation events and also avoid the transgene sugarcane lines exhibited different copies of the cry1Ac silencing associated with complex integration events gene, but harbored the same cassette of the foreign gene [52]. However, it is possible that the low copy number of sequences and exhibited insect resistance [1]. the cry1Ac gene integrated into the sugarcane genome is One interesting finding in the present study is that the ineffective for the improvement of insect resistance, cry1Ac transgenic sugarcane lines with a medium copy while a high copy number of cry1Ac may result in de- number are favorable for the improvement of insect re- terioration of non-objective traits (i.e., yield and quality sistance traits and tend to increase the probability of traits) by the energy consumption [2]. In the present obtaining the transgenic events. This finding is contrary study, the transgene copies of cry1Ac in transgenic sug- to previous studies that suggested that transgenic diploid arcane were variable and random. This illustrated that plants with relatively low copy foreign genes often ex- transformation by particle bombardment produced hibit stable integration and appropriate expression [53]. transgenic sugarcane with complex integration events, The main cause of this difference between the transgenic such as variable copy numbers. The results support diploid plants and sugarcane is probably due to the com- those observations in earlier studies wherein transgene plex ploidy (at least octoploid, chromosome number 120 integration is typically complicated by particle bombard- or more) and large genome (up to 10 Gb) of the modern ment [52–54]. The exogenous cry1Ac gene was ran- sugarcane cultivars. Differences between diploid plants domly inserted into various loci in the sugarcane nuclear and polyploid sugarcane may have influenced the trans- genome. These results are also in agreement with genic traits such as insect resistance. Zhou et al. BMC Plant Biology (2018) 18:342 Page 10 of 14

Another important finding is that different receptor of 138.79 ± 7.80 t/ha, which was almost 1.3-fold higher varieties (FN15 or ROC22) showed different patterns than that of the control sugarcane plants (FN15). with regards to the relationship between cry1Ac protein Furthermore, the yield and theoretical sucrose produc- expression level and cry1Ac gene copy number in trans- tion of two lines (A5 and A6) were equivalent to that of genic sugarcane. For Group I (FN15 background) lines, FN15, while the lines from the receptor ROC22 were a significant linear relationship was found, although one significantly lower than that of ROC22 (Table 4). line (I4) was an exception. The lines with medium copies These results indicated that cry1Ac gene integration (A1–A6) showed high levels of the cry1Ac protein, while increases the genetic diversity of the receptor sugar- lines with low copies (B1, B2, B4, and I2) showed much cane variety. Thus, the transgenic progenies were lower cry1Ac protein amounts. However, for Group II variable and the elite transgenic offspring were bred (ROC22 background), no obvious linear relationship was through a process of selection. observed. For Group II, our findings are inconsistent The current study also provided novel insights into with the dosage effect theory or the GBH theory, which the molecular mechanisms of insect resistance in cry1Ac suggests that the greater the transgene copy number, the transgenic sugarcane lines. The most interesting result is higher its expression level [22, 55, 56]. Transgene copies that the main altered pathways of the transgenic lines can greatly affect expression levels [55, 56]. However, A1 and A5 were enriched in the plant hormone signal the insertion site bias of foreign genes in the genome of transduction, plant-pathogen interaction, benzoxazinoid a receptor and the differences in the rearrangement of biosynthesis, phenylpropanoid biosynthesis, flavonoid foreign genes by the nuclear genome could result from biosynthesis, and linoleic acid metabolism, which are differences in germplasm resources, while the position involved in anti-insect related progresses, while the of the foreign gene integration/insertion site and primary altered pathways of the transgenic line B4 integrity of the expression frame of the foreign gene, can differed from those in A1 and A5. During the process of also have a profound impact on transgene expression plant insect defense, the main function of MAPKs is the [21, 56–58]. Theoretically, the foreign cry1Ac gene can transduction of extracellular stimuli into intracellular re- integrate in the sugarcane genome at virtually any site sponses [59]. The Ca2+ channel and JA signaling path- and almost exclusively at random. The insertion sites of way play a crucial role in the regulation of insect defense the foreign cry1Ac gene may influence transgene expres- responses in plants [40, 41]. This study revealed that the sion such that some integration may occur in higher or DEGs identified by transcriptome analysis were enriched lower transcriptionally active chromatin and the sur- in these defense metabolic pathways, which corroborates rounding endogenous regulatory sequences such as en- the results of previous studies [40, 41]. The JA signaling hancers and inhibitors, or in condensed and even in pathway is considered to be the most important hor- transcriptionally inert chromatin regions [56, 58]. Trans- mone signaling pathway for insect resistance [41]. All of genes in heterochromatic areas such as cry1Ac gene in- the DEGs encoding JAR1, JAZs, and MYC2 in the JA tegration surrounding centromeres, are prone to signaling pathway were up-regulated in the medium- silencing and give rise to reduced and/or variable ex- copy transgenic sugarcane lines A1 and A5, while all of pression. Therefore, it is possible to detect the cry1Ac the DEGs encoding JAZs and MYC2 were down-regu- protein expression in those lines (B1, B2, and B4) with lated in line B4. The differential expression of these very low cry1Ac copies (0~ 1 copies, nearly 0), though JA-related genes may explain the relatively good correl- this is an unusual phenomenon according to the dosage ation between improved insect resistance and medium effect theory. This suggests that a positive correlation copies for the transgenic sugarcane lines A1 and A5. In exists between the cry1Ac protein expression level and addition, the DEGs were mainly involved in plant the copies of cry1Ac gene, if the transgene copies and in- hormone transduction pathways, phenylalanine metab- tegration location are appropriate. olism, glycerophospholipid metabolism, linoleic acid The present study also provides new information on metabolism, flavonoid biosynthesis and diterpenoid the performance of cry1Ac transgenic sugarcane lines. biosynthesis. These metabolic pathways have been The results showed that the cry1Ac transgenic sugarcane classified as resistance induced pathways and play a lines exhibited excellent borer resistance but demon- keyroleinthedefenseagainstinsectattack[60]. strated genotypic differences (Table 2 and Additional file Thus, a synergistic effect of the foreign cry1Ac gene 1: Figure S1). The findings further supported that cry1Ac and the endogenous borer stress-related genes con- gene integration, whether into diploid plants such as O. tributed to improved insect resistance in sugarcane. sativa, or into highly complex polyploid plants such as Further research should aim to investigate the precise sugarcane, can tremendously improve the resistance to function of these defense-related genes. A future borer attack [9]. In terms of agronomic performance, the study focusing on the defense metabolism pathways is transgenic sugarcane line A1 had an average cane yield therefore warranted. Zhou et al. BMC Plant Biology (2018) 18:342 Page 11 of 14

Conclusions 1.0 × 108 to 1.0 × 101 copies per μL were prepared for In this study, foreign cry1Ac gene integration dramatic- assay. Sterile deionized water was used as blank control. ally improved insect resistance in transgenic sugarcane Real-time PCR was performed using an ABI 7500 ther- with medium cry1Ac gene copy numbers. RNA-Seq ana- mal cycler (Foster, USA). The reactions were performed lysis revealed that up/down-regulated DEGs were abun- using final volumes of 25.0 μL, including 12.5 μLof2× dant among the cry1Ac sugarcane lines and the receptor Fast Start Universal ProbeMasterMix, 1.0 μL template variety, and the foreign cry1Ac gene and endogenous DNA (25.0 ng/μL diluted genomic DNA/plasmid), 1.0 μL borer stress-related genes may act synergistically. Three (10 μmol/L) each of the forward and reverse primers and lines (A1, A5, and A6), exhibiting an improved pheno- sterile ddH2O with the following program: 50 °C 2 min; type in terms of yield and quality traits and a lower 95 °C 10 min; 45 cycles of 95 °C 15 s, 60 °C 1 min. Each borer damage ratio were selected and are expected to be sample had three replications. The Ct values were obtained used directly as cultivars or crossing parents for sugar- after the reaction. A standard curve (y = k × X + b) was cane borer resistance breeding in the future. established by plotting the Ct vs. the natural log of the cop- ies. The total cry1Ac copies (10Xt) was calculated by relating Methods the Ct value (Yt) to the corresponding standard curve. Plant materials Then, the single cell copy number of each sample was cal- Two receptor varieties (FN15 and ROC22) and 17 culated using the following formula [17]: copies/genome − cry1Ac transgenic sugarcane lines (Group I, 11 lines =10Xt/[25 n g × 10 9 ×6.02×1023/ (10,000 (M bp) × 106 × from receptor variety FN15: A1, A2, A3, A4, A5, A6, B1, 660)]. The data are presented as mean values with standard B2, B4, I2, and I4; Group II, six lines from receptor var- error (SE), and one-way repeated measures analysis of vari- iety ROC22: D1, D2, D4, K2, K3 and K5) were nursed by ance (ANOVA) was used to test the differences in single the Key Laboratory of Sugarcane Biology and Genetic cell copy number of the cry1Ac gene in the sugarcane sam- Breeding, Ministry of Agriculture/Fujian Agriculture and ples. All of the analyses were conducted in EXCEL 2013 Forestry University, China. All of these transgenic sugar- (Microsoft Inc., Redmond, WA, USA) and SPSS 11.5 statis- cane lines with the foreign cry1Ac gene were derived tical software (SPSS Inc., Chicago, IL, USA). from particle bombardment [18]. Cry1Ac protein expression Estimation of foreign cry1Ac gene copy number and A QuantiPlate™ Kit for Cry1Ab/Cry1Ac (Envirologix, detection of protein expression Inc., USA) was used to measure the cry1Ac protein con- Fresh leaves from three individual plants from the same tent in the transgenic sugarcane leaves according to the line, both for the cry1Ac transgenic lines and the recep- manufacturer’s instructions. Each sample had three rep- tor varieties, were sampled as biological replicates for lications. Cry1Ac standards at concentrations of 0.15625, the estimation of cryAc copies and determination of 0.3125, 0.625, 1.25, 2.5 and 5.0 ppb were used for cali- cry1Ac protein content. Total genomic DNA was ex- bration. OD450 values were measured with a microplate tracted using a CTAB protocol described previously [17]. reader (Biotek, gene, USA). The data are presented as The DNA quality and integrity were assessed based on mean values with SE, and the cry1Ac protein concentra- the A260/A280 ratio and electrophoresis. All of the DNA tion data were analyzed by one-way repeated measures samples were stored at − 20 °C. ANOVA. Percentage data (protein concentration in the sugarcane leaves) were transformed using arcsine Foreign gene cry1Ac copy number [square root (x)] prior to analysis. All of the analyses A quantitative TaqMan real-time PCR method was were conducted in EXCEL 2013 (Microsoft Inc., established to estimate cry1Ac gene copies. Three en- Redmond, WA, USA), SPSS 11.5 statistical software dogenous reference genes with potential low copy num- (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 6 bers, adenosine-5- phosphosulfate reductase (APRT), (GraphPad Prism Software Inc., San Diego, CA, USA). cyclin (CYC), and prolyl-4-hydroxylase (P4H) were cloned and a multi-endogenous reference gene standard Evaluation of stalk borer damage level and phenotypic traits plasmid (p1AcAPC) for transgene copy number identifi- The transgenic sugarcane lines and their receptor var- cation was constructed [2, 61]. This plasmid p1AcAPC ieties were employed for phenotypic trait analysis. The contained not only the three endogenous reference phenotypic traits and stem border damage ratio of these genes APRT, P4H and CYC, but also the foreign gene transgenic sugarcane lines and their receptor were inves- cry1Ac (see the Additional file 4: Figure S4). The primer tigated in field trials carried out at the field station sequences and TaqMan probe sequences for the located at Fujian Agriculture and Forestry University. real-time PCR are listed in Additional file 5: Table S5. In the isolated field in Fuzhou, which had been Tenfold serial dilutions of the plasmid p1AcAPC with approved by the Ministry of Agriculture of China, the Zhou et al. BMC Plant Biology (2018) 18:342 Page 12 of 14

cry1Ac transgenic sugarcane lines and non-transgenic assembly of the clean reads, and functional annotation controls were evaluated from 2013 to 2014 using a ran- were conducted as previously described [28]. Clean reads domized complete block design with three replications. were obtained using a perl script by removing poly-A tails, Each block had three rows, and each row was 8 m in adaptors and contaminants. The clean reads from each length with 1.3 m spacing between the rows. Each block sample were hen merged and de novo assembled by the covered an area of 31.2 m2. All the sugarcane plants Trinity Program as a reference library, due to absence of a were cut into single bud setts and were planted with 15 sugarcane reference genome [28, 62]. Function annotation single bud setts per meter. During the harvest season in was achieved by searching the unigenes against the Nr, January 2014, phenotypic traits were investigated. Stem COG (Clusters of Orthologous Groups of proteins data- diameter (about 1 meter above the base of the stalk), base), KEGG and Swiss-Prot [63]. Gene Ontology (GO) stalk height (from the base of the stalk to the first visible annotation and functional classification were performed dewlap), and Brix values were measured in 10 consecu- using Blast2GO and WEGO, respectively. tive principal stalks in each block. The number of mill- DEGs between the cry1Ac gene “medium-copy” group able stalks and the theoretical cane yield per hectare (n =3)and “low-copy” group (n = 3) were analyzed using were calculated based on the area, number and weight EdgeR [64]. Reads per kilobase per million reads (RPKM) of the stalks, and the theoretical sucrose yield was calcu- values were used to normalize gene expression levels. Genes lated based on the average cane yield and sucrose content with more than 2-fold change (log2FC ≥ 1) with P-value using equations as described by Wang et al. [19]. Twenty ≤0.01 and false discovery rate (FDR) ≤ 0.05 were classed as individual plants in each block were randomly measured DEGs in this study. KEGG enrichment analysis was used to to investigate the stalk borer damage. Stalk borer damage dissect the molecular mechanisms of borer resistance ac- level was calculated according to the percentage of stalks cording to a method described previously [65]. After using damaged by the borer in all of the stalks. The results were the log10(RPKM) values of the DEGs for normalization, a expressed as the mean values ±SE of three replicates. Stat- heat map was performed using the heat map illustrator istical analyses were performed by Microsoft Excel 2013 software HemI 1.0.3.7 (http://hemi.biocuckoo.org/). and SPSS11.5 (SPSS Inc., Chicago, IL, USA) using Stu- dent’s t test and one way ANOVA. Additional files

Transcriptome analysis Additional file 1: Figure S1. Insect-resistant phenotype of the cry1Ac According to the copy number estimation results, three sugarcane and the receptor variety FN15 (CK). (PDF 314 kb) cry1Ac transgenic sugarcane lines and the non-transgenic Additional file 2: Supplementary RNA-Seq data file, including Table S1, Table S2, Table S3, Figure S2 and Figure S3 Table S1. The base pair sugarcane line FN15 (CK) were selected for RNA-Seq. information of high quality clean reads for de novo assembly. Table S2. After inoculating with D. saccharalis, the youngest Assembly statistics of the sugarcane unigenes. Table S3. Unigenes basic fully-expanded leaf (namely + 1 leaf) was collected from annotation statistics with Nr, Swiss-Prot, KOG and KEGG. Figure S2. The length distribution of sugarcane unigenes. Figure S3. Statistics of these four sugarcane lines, which were nursed in the differentially expressed genes. (PDF 56 kb) greenhouse of the Key Laboratory of Sugarcane Biology Additional file 3: Table S4. Differentially expressed unigenes involved and Genetic Breeding, Ministry of Agriculture/Fujian in borer-stress responsive pathway. (XLSX 21 kb) Agriculture and Forestry University, China. Five leaves Additional file 4: Figure S4. Construction of the plasmid pG1AcAPC were collected randomly from five individual plants of (p1AcAPC). (PDF 148 kb) each sugarcane line, and combined as one biological repli- Additional file 5: Table S5. The primer sequences and TaqMan probe sequences for real-time PCR. (DOCX 12 kb) cate. Three biological replicates were assessed for each line. All of the samples were frozen in liquid nitrogen im- mediately after sampling and stored at − 80 °C until total Abbreviations APRT: Adenosine-5- phosphosulfate reductase; CDPK: Calcium-dependent RNA extraction. Total RNA was isolated using the TRIzol protein kinase; CYC: Cyclin; DEGs: Differentially expressed genes; DIMBOA/ Reagent Kit (Invitrogen, Carlsbad, CA, USA) following by DIBOA: Hydroxamic acids; FDR: False discovery rate; JAR1: Jasmonic acid the manufacturer’s instructions. Ten micrograms of high resistant 1; JAZs: Jasmonate ZIM-domain proteins; KEGG: Kyoto encyclopedia of genes and genomes; MAPK: Mitogen-activated protein kinase; MYC2: A quality RNA was used for the RNA-Seq analysis (Illumina basic-helix-loop-helix transcription factor; P4H: Prolyl-4-hydroxylase; HiSeq™ 2500/PE125). PAL: Phenylalanine ammonia lyase; ROS: Reactive oxygen species; Transcriptome sequencing, de novo assembly, evalu- RPKM: Reads per kilobase per million reads ation and functional annotation were performed by Acknowledgements Gene Denovo Co., Guangzhou, China. All of the sequen- Not applicable. cing reads were deposited in the National Center for Biotechnology Information under the Bioproject number Funding This work was funded by National Natural Science Foundation of China PRJNA436063 with the Sequence Read Archive (SRA) (Grant No. 31701491), the Sugar Crop Research System of China (CARS-17), accession number SRP133796. Data filtering, de novo Science and Technology Major Project of Fujian Province (2015NZ0002–2), Zhou et al. BMC Plant Biology (2018) 18:342 Page 13 of 14

the open funding of National Engineering Research Center for Sugarcane, 10. Chen HX, Yang R, Yang W, Zhang L, Camara I, Dong XH, Liu YQ, Shi WP. Fujian Agriculture and Forestry University (Grant No. 2017.6.2) and supported Efficacy of Bt maize producing the Cry1Ac protein against two important by the Scientific Research Foundation of Graduate School of Fujian pests of corn in China. Environ Sci Pollut Res Int. 2016;23:21511–6. Agriculture and Forestry University (Grant No. 1122YB015). Doctoral Scientific 11. Niu L, Tian Z, Liu H, Zhou H, Ma W, Lei C, Chen L. Transgenic Bt cotton Research Start-up Funding of Hunan University of Science and Technology expressing Cry1Ac/Cry2Ab or Cry1Ac/EPSPS does not affect the plant bug (E51761). Adelphocoris suturalis or the pollinating beetle Haptoncus luteolus. Environ Pollut. 2018;234:788–93. Availability of data and materials 12. Valderrama AM, Veasquez N, Rodriguez E, Zapata A, Zaidi MA, Altosaar I, The datasets supporting the conclusions of this manuscript are included Arango R. Resistance to Tecia solanivora (Lepidoptera: Gelechiidae) in three within the article and its additional files. transgenic Andean varieties of potato expressing Bacillus thuringiensis CrylAc protein. J Econ Entomol. 2007;100:172–9. Authors’ contributions 13. Marques LH, Santos AC, Castro BA, Moscardini VF, Rossetto J, Silva OAN, DGZ, LPX and YXQ conceived and designed the experiments. DGZ, XLL and Zobiole LHS, Valverde-Garcia P, Babcock JM, Storer NP, et al. Field evaluation SWG prepared materials. DGZ, XLL, SWG, JLG, YCS, HL, CFW and ZL performed of soybean transgenic event DAS-81419-2 expressing Cry1F and Cry1Ac the experiments. DGZ, XLL, LPX and YXQ analyzed the data. DGZ, XLL, and YXQ proteins for the control of secondary lepidopteran pests in Brazil. Crop wrote the paper. LPX and YXQ revised and approved the final version of the Protect. 2017;96:109–15. manuscript. All authors read and approved the final manuscript. 14. Liu Y, Liu Q, Wang Y, Chen X, Song X, Romeis J, Li Y, Peng Y. Ingestion of Bt corn pollen containing Cry1Ab/2Aj or Cry1Ac does not harm Propylea Ethics approval and consent to participate japonica larvae. Sci Rep. 2016;6:23507. Not applicable. 15. Shabbir MZ, Quan YD, Wang ZY, Bravo A, Soberon M, He KL. Characterization of the Cry1Ah resistance in Asian corn borer and its cross- Consent for publication resistance to other Bacillus thuringiensis toxins. Sci Rep. 2018;8:234. Not applicable. 16. Karthikeyan A, Valarmathi R, Nandini S, Nandhakumar M. Genetically modified crops: insect resistance. Biotechnol. 2012;11:119–26. Competing interests 17. Zhou DG, Guo JL, Xu LP, Gao SW, Lin QL, Wu QB, Wu LG, Que YX. Establishment The authors declare that they have no competing interests. and application of a loop-mediated isothermal amplification (LAMP) system for detection of cry1Ac transgenic sugarcane. Sci Rep. 2014;4:4912. Publisher’sNote 18. Gao S, Yang Y, Wang C, Guo J, Zhou D, Wu Q, Su Y, Xu L, Que Y. Transgenic Springer Nature remains neutral with regard to jurisdictional claims in sugarcane with a cry1Ac gene exhibited better phenotypic traits and published maps and institutional affiliations. enhanced resistance against sugarcane borer. PLoS One. 2016;11:e0153929. 19. Wang WZ, Yang BP, Feng XY, Cao ZY, Feng CL, Wang JG, Xiong GR, Shen Author details LB, Zeng J, Zhao TT. Development and characterization of transgenic 1Key Laboratory of Sugarcane Biology and Genetic Breeding, Fujian sugarcane with insect resistance and herbicide tolerance. Front Plant Sci. Agriculture and Forestry University, Ministry of Agriculture, Fuzhou 350002, 2017;8:1535. Fujian, China. 2Key Laboratory of Ecological Remediation and Safe Utilization 20. Arencibia AD, Vázquez RI, Prieto D, Téllez P, Carmona ER, Coego A, of Heavy Metal-Polluted Soils, Hunan University of Science and Technology, Hernández L, Gustavo A, Selman-Housein G. Transgenic sugarcane plants – School of Life Science, Xiangtan 411201, Hunan, China. resistant to stem borer attack. Mol Breed. 1997;3:247 55. 21. Dai S, Zheng P, Marmey P, Zhang S, Tian W, Chen S, Beachy RN, Received: 15 May 2018 Accepted: 19 November 2018 Fauquet C. Comparative analysis of transgenic rice plants obtained by Agrobacterium-mediated transformation and particle bombardment. Mol Breed. 2001;7:25–33. References 22. Coate JE, Song MJ, Bombarely A, Doyle JJ. Expression-level support for gene 1. Ismail RM. Evaluation of genetically modified sugarcane lines carrying Cry1Ac dosage sensitivity in three Glycine subgenus Glycine polyploids and their – gene using molecular marker techniques. GM Crops Food. 2013;4:58–66. diploid progenitors. New Phytol. 2016;212:1083 93. ’ 2. Zhou DG. Studies on the molecular and biological characteristics of cry1Ac 23. Joyce P, Hermann S, O Connell A, Dinh Q, Shumbe L, Lakshmanan P. Field transgenic sugarcane. PhD Thesis, Fuzhou: Fujian Agriculture and Forestry performance of transgenic sugarcane produced using agrobacterium and – University; 2016. biolistics methods. Plant Biotechnol J. 2014;12:411 24. 3. Islam N, Laksana C, Chanprame S. Agrobacterium-mediated transformation 24. Thomas PE, Kaniewski WK. Agronomic performance of transgenic plants. In: and expression of bt gene in transgenic sugarcane. J ISSAAS. 2016;22:84–95. Foster GD, Taylor SC, editors. Plant virology protocols: from virus isolation to 4. Weng LX, Deng HH, Xu JL, Li Q, Zhang YQ, Jiang ZD, Li Q-W, Chen JW, transgenic resistance. Totowa: Humana Press; 1998. p. 509–18. Zhang LH. Transgenic sugarcane plants expressing high levels of modified 25. Arencibia AD, Carmona ER, Cornide MT, Castiglione S, O'Relly J, Chinea A, cry1Ac provide effective control against stem borers in field trials. Oramas P, Sala F. Somaclonal variation in insect-resistant transgenic Transgenic Res. 2011;20:759–72. sugarcane (Saccharum hybrid) plants produced by cell electroporation. 5. Hoang NV, Furtado A, Botha FC, Simmons BA, Henry RJ. Potential for Transgenic Res. 1999;8:349–60. genetic improvement of sugarcane as a source of biomass for biofuels. 26. Arruda P. Genetically modified sugarcane for bioenergy generation. Curr Front Bioeng Biotechnol. 2015;3:182. Opin Biotechnol. 2012;23:315–22. 6. Basso MF, da Cunha BADB, Ribeiro AP, Martins PK, de Souza WR, de Oliveira 27. Manners JM, Casu RE. Transcriptome analysis and functional genomics of NG, Nakayama TJ, Augusto das Chagas Noqueli Casari R, Santiago TR, sugarcane. Trop Plant Biol. 2011;4:9–21. Vinecky F. Improved genetic transformation of sugarcane (Saccharum spp.) 28. Strickler SR, Bombarely A, Mueller LA. Designing a transcriptome next- embryogenic callus mediated by Agrobacterium tumefaciens. Curr Protoc generation sequencing project for a nonmodel plant species1. Am J Bot. Plant Biol. 2017;2:221–39. 2012;99:257–66. 7. Ming R, Moore PH, Wu KK, D'Hont A, Glaszmann JC, Tew TL. Sugarcane 29. Kristensen C, Morant M, Olsen CE, Ekstrøm CT, Galbraith DW, Møller BL, Bak improvement through breedingand biotechnology. Plant Breed Rev. 2006; S. Metabolic engineering of dhurrin in transgenic Arabidopsis plants with 27:15–118. marginal inadvertent effects on the metabolome and transcriptome. Proc 8. Anunanthini P, Kumar SR, Sathishkumar R. Factors affecting genetic Natl Acad Sci U S A. 2005;102:1779–84. transformation efficiency in sugarcane. In: Mohan C, editor. Sugarcane 30. Krishnamurthy P, Mohanty B, Wijaya E, Lee D-Y, Lim T-M, Lin Q, Xu J, Loh C- biotechnology: challenges and prospects. Cham: Springer International S, Kumar PP. Transcriptomics analysis of salt stress tolerance in the roots of Publishing; 2017. p. 61–73. the mangrove Avicennia officinalis. Sci Rep. 2017;7:10031. 9. Lin N, Mannakkara A, Lin Q, Wang X, Hua H, Lei C, Juratfuentes JL, Ma W. 31. Landis JB, Soltis DE, Soltis PS. Comparative transcriptomic analysis of the Transgenic Bt rice lines producing Cry1Ac, Cry2Aa or Cry1Ca have no detrimental evolution and development of flower size in Saltugilia (Polemoniaceae). effects on Brown Planthopper and Pond Wolf Spider. Sci Rep. 2017;7:1940. BMC Genomics. 2017;18:475. Zhou et al. BMC Plant Biology (2018) 18:342 Page 14 of 14

32. Leyva-Pérez MO, Jiménez-Ruiz J, Gómez-Lama Cabanás C, Valverde- expression stability following biolistic or Agrobacterium-mediated Corredor A, Barroso JB, Luque F, Mercado-Blanco J. Tolerance of olive transformation of sugarcane. In Vitro Cell Dev Biol Plant. 2015;51:603–11. (Olea europaea) cv Frantoio to Verticillium dahliae relies on both basal 55. Azhakanandam K, Mccabe MS, Power JB, Lowe KC, Cocking EC, Davey MR. and pathogen-induced differential transcriptomic responses. New T-DNA transfer, integration, expression and inheritance in rice: effects of Phytol. 2018;217:671–86. plant genotype and Agrobacterium super-virulence. J Plant Physiol. 2000;157: 33. Ricroch AE, Bergé JB, Kuntz M. Evaluation of genetically engineered crops 429–39. using transcriptomic, proteomic, and metabolomic profiling techniques. 56. Tang J, Scarth R, Fristensky B. Effects of genomic position and copy number Plant Physiol. 2011;155:1752–61. of acyl-ACP thioesterase transgenes on the level of the target fatty acids in 34. Jiang Q, Niu F, Sun X, Hu Z, Li X, Ma Y, Zhang H. RNA-seq analysis of Brassica napus L. Mol Breed. 2003;12:71–81. unintended effects in transgenic wheat overexpressing the transcription 57. Vain P, James A, Worland B, Snape W. Transgene behaviour across two factor GmDREB1. Crop J. 2017;5:207–18. generations in a large random population of transgenic rice plants 35. Misra P, Pandey A, Tiwari M, Chandrashekar K, Sidhu OP, Asif MH, produced by particle bombardment. Theor Appl Genet. 2002;105:878. Chakrabarty D, Singh PK, Trivedi PK, Nath P. Modulation of transcriptome 58. Butaye KMJ, Cammue BPA, Delauré SL, De Bolle MFC. Approaches to minimize and metabolome of tobacco by Arabidopsis transcription factor, AtMYB12, variation of transgene expression in plants. Mol Breed. 2005;16:79–91. leads to insect resistance. Plant Physiol. 2010;152:2258–68. 59. Singh IK, Kumar S, Singh S, Singh A. Expression profiling of mitogen- 36. Nietzsche M, Guerra T, Alseekh S, Wiermer M, Sonnewald S, Fernie AR, activated protein kinase genes from chickpea (Cicer arietinum L.) in response Börnke F. STOREKEEPER RELATED 1/G-element binding protein (STKR1) to Helicoverpa armigera, wounding and signaling compounds. J Asia-Pacif interacts with protein kinase SnRK1. Plant Physiol. 2018;176:1773–92. Entomol. 2017;20:942–8. 37. CaiRH,DaiW,ZhangCS,WangY,WuM,ZhaoY,MaQ,XiangY, 60. Chu B, Zhang S, Wang L, Zhu XZ, Luo JY, Wang CY, Lu LM, Cui JJ. Genetic Cheng BJ. The maize WRKY transcription factor ZmWRKY17 negatively regulation of defence responses in cotton to insect herbivores. Aob Plants. regulates salt stress tolerance in transgenic Arabidopsis plants. Planta. 2017;9:10. 2017;246:1215–31. 61. Xue B, Guo J, Que Y, Fu Z, Wu L, Xu L. Selection of suitable endogenous 38. Chung PJ, Jung H, Choi YD, Kim J-K. Genome-wide analyses of direct target reference genes for relative copy number detection in sugarcane. Int J Mol genes of four rice NAC-domain transcription factors involved in drought Sci. 2014;15:8846–62. tolerance. BMC Genomics. 2018;19:40. 62. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis 39. Zhou DG. Molecular identification of cry1Ac transgenic sugarcane and X, Fan L, Raychowdhury R, Zeng Q. Full-length transcriptome assembly from intermediate test evaluation. Master Thesis, Fuzhou: Fujian Agriculture and RNA-Seq data without a reference genome. Nat Biotechnol. 2011;29:644. Forestry University; 2013. 63. Lu X, Kim H, Zhong S, Chen H, Hu Z, Zhou B. De novo transcriptome 40. Wu JQ, Baldwin IT. New insights into plant responses to the attack from assembly for rudimentary leaves in Litchi chinesis Sonn. And identification of insect herbivores. Annu Rev Genet. 2010;44:1–24. differentially expressed genes in response to reactive oxygen species. BMC 41. Schuman MC, Baldwin IT. The layers of plant responses to insect herbivores. Genomics. 2014;15:805. Annu Rev Entomol. 2016;61:373–94. 64. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a bioconductor package for 42. Jouanin L, Bonadé-Bottino M, Girard C, Morrot G, Giband M. Transgenic differential expression analysis of digital gene expression data. – plants for insect resistance. Plant Sci. 1998;131:1–11. Bioinformatics. 2010;26:139 40. 43. Arruda P. Genetically modified sugarcane for bioenergy generation. Curr 65. Xue J, Bao YY, Li BL, Cheng YB, Peng ZY, Liu H, Xu HJ, Zhu ZR, Lou YG, Opin Biotechnol. 2012;23:315-22. Cheng JA. Transcriptome analysis of the brown planthopper Nilaparvata 44. Waclawovsky AJ, Sato PM, Lembke CG, Moore PH, Souza GM. Sugarcane for lugens. PLoS One. 2010;5:e14233. bioenergy production: an assessment of yield and regulation of sucrose content. Plant Biotechnol J. 2010;8:263. 45. Hao W, Altpeter F. Sugarcane ( Saccharum Spp. hybrids). Methods Mol Biol. 2015;1224:307. 46. Sun Y, Joyce PA. Application of droplet digital PCR to determine copy number of endogenous genes and transgenes in sugarcane. Plant Cell Rep. 2017;36:1775–83. 47. Gatehouse JA. Biotechnological prospects for engineering insect-resistant plants. Plant Physiol. 2008;146:881–7. 48. Prado JR, Segers G, Voelker T, Carson D, Dobert R, Phillips J, Cook K, Cornejo C, Monken J, Grapes L. Genetically engineered crops: from idea to product. Annu Rev Plant Biol. 2014;65:769–90. 49. Montero M, Coll A, Nadal A, Messeguer J, Pla M. Only half the transcriptomic differences between resistant genetically modified and conventional rice are associated with the transgene. Plant Biotechnol J. 2011;9:693–702. 50. Bhavani B, Visalakshi M, Rao KV. Integrated management of sugarcane borer complex through mass trapping with pheromone lures and biological controls. Adv Appl Sci. 2017;3:46–50. 51. Yang F, Kerns DL, Brown S, Kurtz R, Dennehy T, Braxton B, Head G, Huang F. Performance and cross-crop resistance of Cry1F-maize selected Spodoptera frugiperda on transgenic Bt cotton: implications for resistance management. Sci Rep. 2016;6:28059. 52. Jackson MA, Anderson DJ, Birch RG. Comparison of Agrobacterium and particle bombardment using whole plasmid or minimal cassette for production of high-expressing, low-copy transgenic plants. Transgenic Res. 2013;22:143–51. 53. Register JC, Peterson DJ, Bell PJ, Bullock WP, Evans IJ, Frame B, Greenland AJ, Higgs NS, Jepson I, Jiao S, Lewnau C, Sillick JM, Wilson HM. Structure and function of selectable and non-selectable transgenes in maize after introduction by particle bombardment. Plant Mol Biol. 1994;25:951–61. 54. Wu H, Awan FS, Vilarinho A, Zeng Q, Kannan B, Phipps T, McCuiston J, Wang W, Caffall K, Altpeter F. Transgene integration complexity and